ngc 6888 3 hours

[alacant]

Hi everyone. It's not often my patience extends beyond its current 2 hour limit, but this target frustrated me enough to have a go for another night. Also an exercise in DSS's group stacking feature; 2 sets each with their own flat frames.

I think I may have reached the limit without false colour filters. There's a bit of detail showing but I doubt any more time would get me much more detail. I wonder...

Thanks for looking and any tips for improvement -particularly processing this beast-  most welcome.

700d on 150-f8: full frame aps-c.

[Dave S]

If you are using DSS groups are you aware that any calibration frames put into the main group tab will get applied across all the other group tabs and will override any calibration frames put into the other groups?  This appears to be an undocumented feature in DSS.  The workaround that seems to be employed most often is to load one light frame (not from the object you are processing) into the main group, but leave it unchecked so it isn't processed and load the lights and calibration frames into the other groups you've set up so that the calibration frames get applied to the correct set of lights.

Might help tighten up the image a bit more.

 

Dave...

[Starlight 1]

That right Dave  and its the only way of doing let say year 1 and year 2 to make 1 image by putting all data from each year in box 1 and box 2 and as you said you can use any image to start with as you only using it to make new boxes.

[Dave S]

4 minutes ago, Starlight 1 said:

That right Dave  and its the only way of doing let say year 1 and year 2 to make 1 image by putting all data from each year in box 1 and box 2 and as you said you can use any image to start with as you only using it to make new boxes.

Ta. Forgot to mention that it is vital when using sets that are separated by such periods of time that they require completely different sets of calibration frames.

Dave...

[alacant]

2 hours ago, Dave S said:

If you are using DSS groups are you aware that any calibration frames put into the main group tab will get applied across all the other group tabs and will override any calibration frames put into the other groups?

No, no idea. Maybe that's good news. With this advice, I think it's worth a reprocess from raw.

Am I correct in thinking that the only frame in the main group be an unchecked light frame?

TIA

[Starlight 1]

11 minutes ago, alacant said:

Am I correct in thinking that the only frame in the main group be an unchecked light frame?

Yes that right its only so you can make a new box , and when you have all day 1data in box 1 you see box 2 turn up so every thing from day 2 go in this box . Dave have good way of putting this.  so its check and uncheck the one frame from main group after you done  let say box 1 and 2

[higgins]

8 hours ago, Dave S said:

If you are using DSS groups are you aware that any calibration frames put into the main group tab will get applied across all the other group tabs and will override any calibration frames put into the other groups?  This appears to be an undocumented feature in DSS.  The workaround that seems to be employed most often is to load one light frame (not from the object you are processing) into the main group, but leave it unchecked so it isn't processed and load the lights and calibration frames into the other groups you've set up so that the calibration frames get applied to the correct set of lights.

Might help tighten up the image a bit more.

 

Dave...

If the bias frames are common to all groups putting them in the main group alone is another way.

 

[alacant]

1 hour ago, higgins said:

bias frames

Yes, thanks. FWIW, it would not let me add bias frames in more than one group anyway; it seems you can only have one set and they have to be in main to have any meaning. Anyone?

[ollypenrice]

I've never heard of false colour filters...  Narrowband filters specify with absolute precision what colour they are. You are shooting with twice as much green as red or blue. is this true colour imaging? If so, why?

lly

[bobro]

Hope you don't mind me saying but I think pushing the colour a bit too much (in StarTools?) has lost detail, especially in the lower part of the nebula. Otherwise looks great!

[alacant]

10 hours ago, ollypenrice said:

I've never heard of false colour filters

? Possibly the best way to describe it is to take a daylight snap through the filters which make space shots look yellow and blue. Grass would probably be purple! HTH.

[alacant]

This is with -I think- the DSS groups properly organised.

main: offset

1: light + flat session 1

2: light and flat session 2

If anything, worse:. I probably overdid the GIMP denoise. As always, no free lunch. So, no more excuses, just need more sessions...

Cheers and clear skies everyone and thanks so much for your comments.

[alacant]

8 hours ago, bobro said:

(in StarTools?) has lost detail, especially in the lower part of the nebula.

Hi. No. This was one of my GIMP only sessions and yes, I agree. The extremes of the nebula aren't equal in brightness. Not sure about in reality. Cheers and clear skies.

[ollypenrice]

1 hour ago, alacant said:

? Possibly the best way to describe it is to take a daylight snap through the filters which make space shots look yellow and blue. Grass would probably be purple! HTH.

The red, green and blue filters used with mono CCD cameras are either the same, or essentially the same, as those used to make monochrome CMOS chips in DSLRs give life-like colour. (DSLRs use absorption filters while the better astro filters are interferometric, but the result is more or less the same.)

Narrowband filters specify exactly what colour they pass and with what bandwidth. Equally you can just look through them in sunlight. There is nothing false about this. If a narrowband imager chooses to map the deep red of Ha to blue then it is the process, not the filter, which creates false colour. Equally, though, an imager can use a narrowband filter to enhance the true colour channel in which it lies, so Ha to red and OIII to the blue-green border. The imager can also compare the post-enhancement colour balance with the natural LRGB to ensure that no radical colour change has occurred.

The trouble with using a term like 'false colour filters' is that it might perpetuate the fallacy that all deep sky images have colour drummed up at the whim of the imager. They don't. This only describes colour mapping in narrowband.

So there is no such thing as a false colour filter.

Olly

[Starlight 1]

Nice image , on my computer the sky black is to black .

[alacant]

16 minutes ago, ollypenrice said:

So there is no such thing as a false colour filter.

OK, ok.... I'll say narrow instead!

17 minutes ago, Starlight 1 said:

the sky black is to black .

Yeah, I think I took the slider too far but still the histogram looks ok. Just. Dunno...

 

[ollypenrice]

1 hour ago, alacant said:

 

 

Don't worry about the histogram, just measure the background sky brightness.

Olly

[alacant]

1 hour ago, ollypenrice said:

measure the background sky brightness

Hi. OK, thanks but not sure how. Any pointers?

[ollypenrice]

In PS I just use the eyedropper set to Colour Sampler and 3x3 or 5x5 average. It gives you a readout per colour channel, which is useful. I generally like 23/23/23 in RGB. Quite often, if I'm processing with guests, they want it darker! 

I'm sure most processing programmes will let you mouse over parts of the image to see the brightness values.

Olly

[alacant]

3 hours ago, ollypenrice said:

just use the eyedropper

Ah yes, OK. GIMP has radius so with radius 2 I'm getting values of 18-26 in the black bits. I'm not sure whether this FOV has any black bits though. What do you do? Take the slider back again? Or maybe my stretch was wrong in the first place. There's less detail in the second effort...

[ollypenrice]

2 hours ago, alacant said:

Ah yes, OK. GIMP has radius so with radius 2 I'm getting values of 18-26 in the black bits. I'm not sure whether this FOV has any black bits though. What do you do? Take the slider back again? Or maybe my stretch was wrong in the first place. There's less detail in the second effort...

In RGB I think you can consider parts of this image to be 'background sky.' This wouldn't apply in H alpha though.

Normally the thin flat line to the left of the histogram pedestal is made up both of faint signal and of noise. During the levels-and-curves stretching routine I would leave plenty of it unclipped. Get to the full stretch before doing any hard clipping. If there is a huge flat line after a stretch then, yes, clip some of it back but be very conservative at this stage. For me the last black point adjustment is literally the last processing operation. I leave plenty of room on the left because all sorts of processing actions nibble into it along the way and when it's gone it's gone...

Olly

[steppenwolf]

17 hours ago, ollypenrice said:

Normally the thin flat line to the left of the histogram pedestal is made up both of faint signal and of noise. During the levels-and-curves stretching routine I would leave plenty of it unclipped. Get to the full stretch before doing any hard clipping.

Absolutely agree with this - MaxIm DL automatically adds a 100 point pedestal to help avoid clipping. When I import FITS files straight into PhotoShop using FITS Liberator, I use a Log (x) stretch for the same reason - you can't have too much adjustment in this part of the histogram if you want to set your backgrounds to a realistic level.

[kirkster501]

I really like the original picture.